Shap

When you need to explain why your ML model made a specific prediction, this gets you SHAP values and visualizations without digging through documentation. It covers the explainer selection decision tree (TreeExplainer for XGBoost and friends, DeepExplainer for neural nets, KernelExplainer for black boxes), then walks through the standard plots: waterfall for individual predictions, beeswarm for global importance, scatter for feature relationships. The workflow coverage is actually useful, especially the debugging and fairness analysis patterns that go beyond just generating plots. Works with pretty much any model type. The main value is having all the explainer types and plot options laid out clearly so you can pick the right tool without trial and error.

skills.sh installs: 373
GitHub stars: 33.2K
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add davila7/claude-code-templates --skill shap --agent claude-code

Installs into .claude/skills of the current project.

Files

SHAP (SHapley Additive exPlanations)

Overview

SHAP is a unified approach to explain machine learning model outputs using Shapley values from cooperative game theory. This skill provides comprehensive guidance for:

Computing SHAP values for any model type
Creating visualizations to understand feature importance
Debugging and validating model behavior
Analyzing fairness and bias
Implementing explainable AI in production

SHAP works with all model types: tree-based models (XGBoost, LightGBM, CatBoost, Random Forest), deep learning models (TensorFlow, PyTorch, Keras), linear models, and black-box models.